Professionals: Designing Challenging AI Prompts
Terac · Remote · United States ·
COMPENSATION
€•• – ••k
Listed on the employer's job board (Ashby)
- One-time paid study
- $20 compensation
- Fully remote
- Requires ChatGPT account
- Screen recording required
This role involves participating in a paid research study to identify experts capable of designing demanding AI tasks. You will translate a real-world workflow into a prompt, test it against ChatGPT to find failure points, and create a grading rubric.
The process is screen-recorded to assess your reasoning. Successful participants may be considered for long-term evaluator opportunities with Terac.
What you'll do
- Convert a familiar professional or personal workflow into a complex AI prompt
- Test the prompt in ChatGPT to identify where the model fails
- Refine the prompt to increase difficulty if the AI succeeds initially
- Write a clear rubric for strangers to grade AI performance on the task
- Share screen, camera, and microphone during the entire process
What you'll need
- Deep knowledge of a specific professional or personal workflow
- Ability to rapidly evaluate AI output accuracy and quality
- Access to a laptop or desktop computer
- Active ChatGPT account
- Comfortable being recorded while solving complex problems
Stack
- ChatGPT
How you'll work
- One-hour time commitment
- Remote participation
- Screen, camera, and audio recording required
What you get
- $20 one-time payment
- Potential consideration for long-term evaluator bench
Hiring process
- Start the paid interview directly via the provided link